CN111784065A - Oil well productivity intelligent prediction method based on grey correlation - Google Patents

Oil well productivity intelligent prediction method based on grey correlation Download PDF

Info

Publication number
CN111784065A
CN111784065A CN202010655166.2A CN202010655166A CN111784065A CN 111784065 A CN111784065 A CN 111784065A CN 202010655166 A CN202010655166 A CN 202010655166A CN 111784065 A CN111784065 A CN 111784065A
Authority
CN
China
Prior art keywords
curve
correlation
capacity
curves
logging
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010655166.2A
Other languages
Chinese (zh)
Other versions
CN111784065B (en
Inventor
付长凤
柴子威
韩连福
刘兴斌
黄赛鹏
刘辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dongying Ruigang Pipeline Engineering Co ltd
Original Assignee
Northeast Petroleum University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northeast Petroleum University filed Critical Northeast Petroleum University
Priority to CN202010655166.2A priority Critical patent/CN111784065B/en
Publication of CN111784065A publication Critical patent/CN111784065A/en
Application granted granted Critical
Publication of CN111784065B publication Critical patent/CN111784065B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/067—Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Primary Health Care (AREA)
  • Mining & Mineral Resources (AREA)
  • Educational Administration (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Animal Husbandry (AREA)
  • Agronomy & Crop Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention belongs to the technical field of petroleum engineering and measurement, and particularly relates to an oil well productivity intelligent prediction method based on ash correlation, which comprises the following steps: 1. selecting a multiplied by N logging curves as template curves in a certain block, wherein a is the number of oil well productivity classification categories, and N is the number of each type of productivity curves; 2. digitizing all the sample curves, and digitizing the well logging curves to be determined; 3. establishing a variable weight grey correlation model for judging the correlation of the logging curve form; 4. determining a weight coefficient of a variable weight gray correlation model determined by the correlation of the well logging curve form; 5. and judging the correlation of the shapes of the logging curves to be tested, and giving out productivity prediction. The prediction method solves the problem that productivity prediction is inaccurate due to the fact that a logging curve is not objectively identified by a manual reading method and a large amount of data is needed for identifying the logging curve based on a neural network big data method.

Description

Oil well productivity intelligent prediction method based on grey correlation
The technical field is as follows:
the invention belongs to the technical field of petroleum engineering and measurement, and particularly relates to an oil well productivity intelligent prediction method based on ash correlation.
Background art:
factors such as geological conditions and crude oil physical properties are important parameters influencing the oil well production performance, and the shape of a logging curve exactly reflects key factors such as the geological conditions of the oil well. The traditional method for predicting the oil well productivity by utilizing the logging curve form mainly comprises an artificial reading method and a big data method based on a neural network, has many defects and shortcomings, and cannot provide effective support for the subsequent oil field production work. The manual reading method is easily influenced by subjective factors. The big data method based on the neural network requires that the cardinality of training samples is large, the data is often hundreds of thousands of samples, and the number of logging curve samples is often not more than 200, so that the identification effect of the method is poor.
The most common gray correlation method is the traditional Duncan gray correlation method, the algorithm principle is simple, the calculation speed is high, and the like, but the method is only used for calculating the gray correlation degree among one dimensional group and is not suitable for calculating the gray correlation degree among two-dimensional image matrixes. The improved grouped sample gray correlation method can calculate the gray correlation degree between two-dimensional image matrixes, but the calculation result of the method only reflects the distance correlation of the two matrixes and cannot reflect the slope correlation of the two matrixes. The improved absolute gray correlation method can reflect the slope correlation of two image matrixes but cannot reflect the distance correlation of the two image matrixes, so that both the grouped sample gray correlation method and the absolute gray correlation method cannot be directly applied to the judgment of the correlation of the well logging curve form.
The invention content is as follows:
the invention aims to solve the problem that productivity prediction is inaccurate due to the fact that a manual reading method is not objective in identifying a logging curve and a neural network big data method is used for identifying the logging curve and a large amount of data is needed, and therefore the logging curve is not accurately identified. Aiming at the problem, in order to overcome the defects of a manual reading method and a neural network big data method-based logging curve identification method, so as to accurately predict the productivity of the oil well, an oil well productivity intelligent prediction method based on grey correlation is provided. The core problem of the logging curve form identification is to judge the form correlation of a curve to be detected and a known sample curve, and the grey correlation is a small sample data processing method and meets the requirement of the small sample problem of the logging curve form identification. Grey correlations may determine the relevance of the curves. And the model is used for identifying the form of the logging curve and predicting the productivity of the oil well, so that the prediction precision is improved.
The technical scheme adopted by the invention is as follows: an oil well productivity intelligent prediction method based on ash correlation comprises the following steps:
the method comprises the following steps: selecting a multiplied by N logging curves as template curves in a certain block, wherein a is the number of oil well productivity classification categories, and N is the number of each type of productivity curves;
step two: digitizing all the sample curves, and digitizing the well logging curves to be determined;
step three: establishing a variable weight grey correlation model for judging the correlation of the logging curve form;
step four: determining a weight coefficient of a variable weight gray correlation model determined by the correlation of the well logging curve form;
step five: judging the correlation of the form of the logging curve to be tested, and giving out productivity prediction;
further, the method for digitizing the logging curve in the second step comprises: and (4) carrying out curve digitization by using neurolog software, and carrying out curve digitization by using the neurolog software.
Further, in the third step, the variable weight gray correlation model for determining the correlation between the well logging curve forms is established, where u is the oil well productivity classification serial number and the curve to be measured and the u is the variable weight gray correlation of the oil well productivity curve group, where u is 1,2, 3.
Figure BDA0002576461140000021
Wherein i represents a curve serial number of the u-type capacity model curve, wherein i is 1,2,3b <u>(x0,xi) The variable weight grey correlation degree g of the ith curve in the u type of the logging curve to be measured and the sample logging curveb <u>(x0,xi) Composed of groups of sample gray associations and absolute gray associations, gb <u>(x0,xi) The expression is as follows:
gb <u>(x0,xi)＝αgz <u>(x0,xi)+βgd <u>(x0,xi)
wherein α is a group of sample weight coefficients of the gray correlation model for well logging curve form correlation determination, β is an absolute weight coefficient of the gray correlation model for well logging curve form correlation determination, and the group of sample gray correlation degrees gz <u>(x0,xi) And an improved absolute grey scale degree gd <u>(x0,xi) The expression of (a) is as follows:
Figure BDA0002576461140000031
Figure BDA0002576461140000032
where m is the number of rows of the matrix generated after the curve is digitized, j is the serial number of the rows of the matrix generated after the curve is digitized, where j is 1,2,3,.., m.n is the number of columns of the matrix generated after the curve is digitized, and k is the serial number of the columns of the matrix generated after the curve is digitized, where k is 1,2,30i(x0j(k),xij(k) Represents the correlation between the kth line data of the jth row in the matrix data generated by the logging curve to be tested and the kth line data of the jth row in the matrix data generated by the logging curve of the ith sample plate under the productivity classificationDegree, omega (x)0j(k +1)) represents the difference between the (k +1) th column data of the jth row and the (k) th column data of the jth row in the matrix generated by the logging curve to be tested, and omega (x)ij(k +1)) represents the difference between the (k +1) th column data of the jth row and the kth column data of the jth row in the matrix generated by the ith sample plate well logging curve under the capacity classification, and the expression is as follows:
Figure BDA0002576461140000041
Ω(x0j(k+1))＝x0j(k+1)-x0j(k)
Ω(xij(k+1))＝xij(k+1)-xij(k)
in the formula, x0j(k) For the data value, x, of the jth row and kth column of the matrix generated after the well logging curve to be measured is curved0j(k +1) is a data value of a jth row and a kth +1 column of a matrix generated after the well logging curve to be detected is curved, and xij(k) For the data value, x, of the jth row and the kth column of the matrix generated after the logging curve of the ith sample plate under the capacity classification is curvedij(k +1) is the data value of the jth row and the k +1 th column of the matrix generated after the ith sample plate well logging curve under the capacity classification is curved.
Further, in the fourth step, the method for calculating the weights α and β of the variable weight gray correlation model for determining the correlation of the log shape is as follows:
Figure BDA0002576461140000042
Figure BDA0002576461140000043
in the formula, gz *(u) average of the gray levels of the set of samples of the last curve in the u-th capacity classification and the other N-1 curves in the capacity classification, gd *(u) is the average value of the improved absolute gray correlation degree of the last curve in the u-th capacity classification and other N-1 curves in the capacity classification, and the calculation method is as follows:
Figure BDA0002576461140000051
Figure BDA0002576461140000052
in the formula, gz **(u, i) group sample gray correlation of the last curve in the u-th capacity classification and the ith curve in the capacity classification, gd **(u, i) is the improved absolute gray correlation degree of the last curve in the u-th capacity classification and the ith curve in the capacity classification, and the calculation method is the same as the g in the third stepz <u>(x0,xi) And gd <u>(x0,xi)；
Further, in the fifth step, the correlation of the form of the logging curve to be measured is determined, and the productivity prediction is given, and the specific operation method is as follows: respectively calculating the grey correlation degree g of the variable weight values of the curve to be measured and the production capacity curve group of the u-th oil wellb <u>And the capacity classification corresponding to the maximum grey correlation degree is the capacity of the curve to be measured, so that the capacity prediction is completed.
The invention has the beneficial effects that: the method solves the problems that the identification of the logging curve by a manual reading method is not objective, and the identification of the logging curve based on a neural network big data method requires a large amount of data, so that the identification of the logging curve is not accurate enough, and the productivity prediction is not accurate. The advantages are as follows:
1) the grey correlation calculation model related by the method overcomes the defects of the traditional Deng correlation, the absolute correlation and the grouped sample correlation, and can calculate the correlation degree between the well logging curves more accurately; .
2) The method for calculating the weight is a comprehensive and objective weighting method which determines the weight according to different proportions of distance correlation and slope correlation among curves of the same category and is a method for calculating the weight;
3) the oil well productivity intelligent prediction method based on gray correlation provided by the method is small in required data quantity, and distance correlation and slope correlation between curves are more suitable for research of a logging curve form method compared with a traditional neural network deep learning algorithm.
Description of the drawings:
FIG. 1 is a flowchart of a log pattern correlation determination method based on gray correlation according to an embodiment;
FIG. 2 is a graph of a prototype log with the productivity of "Medium" for the well A in the first example;
FIG. 3 is a graph of a prototype log with "Medium" productivity for well B in the first example;
FIG. 4 is a graph of a prototype log with a capacity of "Medium" for the well C of the first embodiment;
FIG. 5 is a graph of a prototype log with a productivity of "Medium" for well D in the first example;
FIG. 6 is a graph of a prototype log with "medium" capacity for the E well in the first example;
FIG. 7 is a graph of a prototype log with capacity "Medium" for the first example F well;
FIG. 8 is a log of a pending determination in accordance with one embodiment;
FIG. 9 is a graph of error comparisons between the gray correlation analysis of the weighted values in the first embodiment and the gray correlation analysis of the grouped samples and the absolute gray correlation analysis.
The specific implementation mode is as follows:
example one
Referring to the figures, the method for intelligently predicting the oil well productivity based on ash correlation comprises the following steps:
the method comprises the following steps: selecting 3 types of well logging curve groups with known productivity as template curves in a certain block, wherein the productivity of the well logging curve groups respectively represents high, medium and low, and 6 template curves are selected in each type;
step two: digitizing all the sample curves, and digitizing the well logging curves to be determined;
the specific operation method comprises the following steps:
(1) importing the well logging curve image into neurolog software;
(2) and the abscissa and the ordinate of the given curve are depth, and the abscissa is the value of the logging curve. Respectively selecting a starting depth, an ending depth, a minimum curve value and a maximum curve value on the image;
(3) selecting the name of a digitalized curve, pressing a cursor key of a digitizer to draw the curve, and collecting curve data;
(4) and saving the curve acquisition data in a lass format at a sampling interval of 0.125 meters. (5) Checking and checking the generated data to ensure the correctness of the data.
Step three: establishing a variable weight grey correlation model for judging the correlation of the logging curve form;
the method for calculating the variable weight gray correlation degree of the curve to be measured and the u-th (u is 1,2,3) oil well productivity curve group comprises the following steps:
Figure BDA0002576461140000071
wherein i is 1,2,3b <u>(x0,xi) The grey correlation degree of the variable weight value of the ith curve in the u type of the logging curve to be measured and the sample logging curve is obtained;
gb <u>(x0,xi) The variable weight grey correlation degree g of the ith curve in the u type of the logging curve to be measured and the sample logging curveb <u>(x0,xi) Is composed of groups of sample gray associations and absolute gray associations, gb <u>(x0,xi) The expression is as follows:
gb <u>(x0,xi)＝αgz <u>(x0,xi)+βgd <u>(x0,xi)
wherein α is the group of sample weight coefficients of the gray correlation model for log pattern correlation determination, β is the absolute weight coefficient of the gray correlation model for log pattern correlation determination, gz <u>(x0,xi) Gray correlation for grouped samples, gd <u>(x0,xi) The method is an improved absolute grey correlation degree;
grey correlation g for grouped samplesz <u>(x0,xi) And an improved absolute grey scale degree gd <u>(x0,xi) The expression of (a) is as follows:
Figure BDA0002576461140000081
Figure BDA0002576461140000082
ζ0i(x0j(k),xij(k))、Ω(x0j(k +1)) and Ω (x)ij(k +1) represents a method:
Figure BDA0002576461140000083
Ω(x0j(k+1))＝x0j(k+1)-x0j(k)
Ω(xij(k+1))＝xij(k+1)-xij(k)
in the formula, x0j(k) For the data value, x, of the jth row and kth column of the matrix generated after the well logging curve to be measured is curved0j(k +1) is a data value of a jth row and a kth +1 column of a matrix generated after the well logging curve to be detected is curved, and xij(k) For the data value, x, of the jth row and the kth column of the matrix generated after the logging curve of the ith sample plate under the capacity classification is curvedij(k +1) is the data value of the jth row and the k +1 th column of the matrix generated after the ith sample plate well logging curve under the capacity classification is curved.
Step four: determining a weight coefficient of a variable weight gray correlation model determined by the correlation of the well logging curve form;
the calculation method of the weight alpha and the weight beta of the variable weight gray correlation model for the logging curve form correlation determination is as follows:
Figure BDA0002576461140000091
Figure BDA0002576461140000092
wherein u is 1,2,3, gz *(u) average of the gray levels of the set of samples of the last curve in the u-th capacity classification and the other N-1 curves in the capacity classification, gd *(u) is the average value of the improved absolute gray correlation degree of the last curve in the u-th capacity classification and other 5 curves in the capacity classification, and the calculation method is as follows:
Figure BDA0002576461140000093
Figure BDA0002576461140000094
wherein i is 1,2,3z **(u, i) group sample gray correlation of the last curve in the u-th capacity classification and the ith curve in the capacity classification, gd **(u, i) is the improved absolute gray correlation degree of the last curve in the u-th capacity classification and the ith curve in the capacity classification, and the calculation method is the same as the g in the third stepz <u>(x0,xi) And gd <u>(x0,xi)。
Step five: judging the correlation of the form of the logging curve to be tested, and giving out productivity prediction;
judging the correlation of the form of the logging curve to be tested, and giving out productivity prediction, wherein the specific operation method comprises the following steps: respectively calculating the grey correlation degree g of the variable weight values of the curve to be measured and the u-th (u is 1,2 and 3) oil well productivity curve groupb <u>And the capacity classification corresponding to the maximum grey correlation degree is the capacity of the curve to be measured, so that the capacity prediction is completed.
In order to verify the accuracy of a logging curve form correlation judgment method algorithm based on grey correlation in the method, the method is applied to a simulated well, 3 types of logging curve groups with known productivity are selected as template curves, the productivity of the logging curve groups respectively represents high, medium and low, and 6 template curves are selected for each type. As shown in fig. 2 to 7. And taking the A-F well logging curves as 6 sample plate curves with the productivity of 'medium', digitizing the curves according to the second step, and establishing a logging curve form correlation judgment model based on gray correlation as shown in the third step. And according to the fourth step, calculating the weights alpha and beta of the variable weight gray correlation model determined by the logging curve form correlation to be 0.9065 and 0.0935 respectively. And judging the correlation of the shapes of the logging curves to be measured according to the fifth step, and finally calculating the gray correlation degrees of the variable weight values of the curves to be measured and three groups of sample curves of different classifications as shown in the figure 8 as follows: 0.4520, 0.6012 and 0.2195, and finally determining that the capacity of the curve to be measured is 'middle'. And according to the actual measurement condition, the prediction is accurate.
Fig. 9 compares the prediction accuracy of the gray-correlated analysis model based on the variable weight value with the grouped sample gray-correlated model and the improved absolute gray-correlated model, and it can be seen that the prediction accuracy of the gray-correlated analysis model based on the variable weight value is higher when the number of curve bars of each type of energy production sample plate is selected sufficiently. According to the analysis of the experimental synthesis, the prediction accuracy rate of the grey correlation analysis model based on the variable weight value is 85.06%, which is higher than 81.97% of the grey correlation model of the grouped samples and 60.65% of the improved absolute grey correlation model.
According to the method, the oil well productivity is predicted by adopting an oil well productivity intelligent prediction method based on grey correlation, the defects that the previous productivity prediction needs a large amount of data and the subjectivity of a manual judgment method are overcome, and the precision measurement precision of the oil well productivity prediction is improved.

Claims (5)

1. An oil well productivity intelligent prediction method based on ash correlation is characterized by comprising the following steps: the intelligent prediction method for the oil well productivity comprises the following steps:
the method comprises the following steps: selecting a multiplied by N logging curves as template curves in a certain block, wherein a is the number of oil well productivity classification categories, and N is the number of each type of productivity curves;
step two: digitizing all the sample curves, and digitizing the well logging curves to be determined;
step three: establishing a variable weight grey correlation model for judging the correlation of the logging curve form;
step four: determining a weight coefficient of a variable weight gray correlation model determined by the correlation of the well logging curve form;
step five: and judging the correlation of the shapes of the logging curves to be tested, and giving out productivity prediction.
2. The intelligent oil well capacity prediction method based on ash correlation as claimed in claim 1, wherein: the curves were digitized using neurolog software.
3. The intelligent oil well capacity prediction method based on ash correlation as claimed in claim 1, wherein: in the third step, the curve to be measured and the u-th class are the oil well productivity classification serial numbers, wherein u is 1,2, 3.
Figure FDA0002576461130000011
Wherein i represents a curve serial number of the u-type capacity model curve, wherein i is 1,2,3b <u>(x0,xi) The variable weight grey correlation degree g of the ith curve in the u type of the logging curve to be measured and the sample logging curveb <u>(x0,xi) Composed of groups of sample gray associations and absolute gray associations, gb <u>(x0,xi) The expression is as follows:
gb <u>(x0,xi)＝αgz <u>(x0,xi)+βgd <u>(x0,xi)
wherein α is the group sample weight coefficient of the gray correlation model for the determination of the correlation of the well log shape, β is the weight coefficient of the gray correlation modelThe absolute weight coefficient of a gray correlation model for well curve form correlation determination and the gray correlation degree g of a group of samplesz <u>(x0,xi) And an improved absolute grey scale degree gd <u>(x0,xi) The expression of (a) is as follows:
Figure FDA0002576461130000021
Figure FDA0002576461130000022
where m is the number of rows of the matrix generated after the curve is digitized, j is the serial number of the rows of the matrix generated after the curve is digitized, where j is 1,2,3,.., m.n is the number of columns of the matrix generated after the curve is digitized, and k is the serial number of the columns of the matrix generated after the curve is digitized, where k is 1,2,30i(x0j(k),xij(k) Represents the correlation degree between the kth line data of the jth row in the matrix data generated by the logging curve to be tested and the kth line data of the jth row in the matrix data generated by the logging curve of the ith sample plate under the productivity classification, and is omega (x)0j(k +1)) represents the difference between the (k +1) th column data of the jth row and the (k) th column data of the jth row in the matrix generated by the logging curve to be tested, and omega (x)ij(k +1)) represents the difference between the (k +1) th column data of the jth row and the kth column data of the jth row in the matrix generated by the ith sample plate well logging curve under the capacity classification, and the expression is as follows:
Figure FDA0002576461130000023
Ω(x0j(k+1))＝x0j(k+1)-x0j(k)
Ω(xij(k+1))＝xij(k+1)-xij(k)
in the formula, x0j(k) For the data value, x, of the jth row and kth column of the matrix generated after the well logging curve to be measured is curved0j(k +1) is a data value of a jth row and a kth +1 column of a matrix generated after the well logging curve to be detected is curved, and xij(k) For the data value, x, of the jth row and the kth column of the matrix generated after the logging curve of the ith sample plate under the capacity classification is curvedij(k +1) is the data value of the jth row and the k +1 th column of the matrix generated after the ith sample plate well logging curve under the capacity classification is curved.
4. The intelligent oil well capacity prediction method based on ash correlation as claimed in claim 1, wherein: the method for calculating the weights alpha and beta of the variable weight gray correlation model for judging the correlation of the well logging curve form comprises the following steps:
Figure FDA0002576461130000031
Figure FDA0002576461130000032
in the formula, gz *(u) average of the gray levels of the set of samples of the last curve in the u-th capacity classification and the other N-1 curves in the capacity classification, gd *(u) is the average value of the improved absolute gray correlation degree of the last curve in the u-th capacity classification and other N-1 curves in the capacity classification, and the calculation method is as follows:
Figure FDA0002576461130000041
Figure FDA0002576461130000042
in the formula, gz **(u, i) group sample gray correlation of the last curve in the u-th capacity classification and the ith curve in the capacity classification, gd **(u, i) is the improved absolute gray correlation degree of the last curve in the u-th capacity classification and the ith curve in the capacity classification, and the calculation method is the same as the g in the third stepz <u>(x0,xi) And gd <u>(x0,xi)。
5. The intelligent oil well capacity prediction method based on ash correlation as claimed in claim 1, wherein: the capacity prediction method in the fifth step is as follows:
respectively calculating the grey correlation degree g of the variable weight values of the curve to be measured and the production capacity curve group of the u-th oil wellb <u>And the capacity classification corresponding to the maximum grey correlation degree is the capacity of the curve to be measured, so that the capacity prediction is completed.
CN202010655166.2A 2020-07-09 2020-07-09 Oil well productivity intelligent prediction method based on grey correlation Active CN111784065B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010655166.2A CN111784065B (en) 2020-07-09 2020-07-09 Oil well productivity intelligent prediction method based on grey correlation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010655166.2A CN111784065B (en) 2020-07-09 2020-07-09 Oil well productivity intelligent prediction method based on grey correlation

Publications (2)

Publication Number Publication Date
CN111784065A true CN111784065A (en) 2020-10-16
CN111784065B CN111784065B (en) 2021-04-16

Family

ID=72758540

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010655166.2A Active CN111784065B (en) 2020-07-09 2020-07-09 Oil well productivity intelligent prediction method based on grey correlation

Country Status (1)

Country Link
CN (1) CN111784065B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112949053A (en) * 2021-02-20 2021-06-11 东北石油大学 Dynamic prediction method for water injection rate of water injection oil well
CN114021804A (en) * 2021-11-02 2022-02-08 东北石油大学 Construction method of fault-lithology oil and gas reservoir oil and gas reserve prediction model
CN115576028A (en) * 2022-12-01 2023-01-06 武汉盛华伟业科技股份有限公司 Geological feature layer prediction method and system based on support vector machine

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6640896B1 (en) * 1999-07-15 2003-11-04 China Petroleum & Chemical Corporation Mechanical oil recovery method and system with a sucker rod pump
CN104564006A (en) * 2015-02-04 2015-04-29 中国海洋石油总公司 Hypotonic gas well fracturing water-producing capacity judgment method
CN104778378A (en) * 2015-05-05 2015-07-15 中国石油大学(华东) Method for analyzing influence factors of production decline of oil and gas field
CN106437691A (en) * 2016-08-05 2017-02-22 中国石油集团长城钻探工程有限公司录井公司 Low gas-oil-ratio oil reservoir gas logging evaluation method
CN109975189A (en) * 2017-12-28 2019-07-05 中国石油天然气股份有限公司 Method and device for predicting productivity of pore type sandstone reservoir
CN111144542A (en) * 2019-12-26 2020-05-12 中国石油大学(北京) Oil well productivity prediction method, device and equipment
WO2020101128A1 (en) * 2018-11-15 2020-05-22 한국지질자원연구원 Method for predicting shale gas production by using deep learning

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6640896B1 (en) * 1999-07-15 2003-11-04 China Petroleum & Chemical Corporation Mechanical oil recovery method and system with a sucker rod pump
CN104564006A (en) * 2015-02-04 2015-04-29 中国海洋石油总公司 Hypotonic gas well fracturing water-producing capacity judgment method
CN104778378A (en) * 2015-05-05 2015-07-15 中国石油大学(华东) Method for analyzing influence factors of production decline of oil and gas field
CN106437691A (en) * 2016-08-05 2017-02-22 中国石油集团长城钻探工程有限公司录井公司 Low gas-oil-ratio oil reservoir gas logging evaluation method
CN109975189A (en) * 2017-12-28 2019-07-05 中国石油天然气股份有限公司 Method and device for predicting productivity of pore type sandstone reservoir
WO2020101128A1 (en) * 2018-11-15 2020-05-22 한국지질자원연구원 Method for predicting shale gas production by using deep learning
CN111144542A (en) * 2019-12-26 2020-05-12 中国石油大学(北京) Oil well productivity prediction method, device and equipment

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
HUAINIAN ZHANG: "EMD-based Gray Association Algorithm for Group Ballistic Target", 《IEEE》 *
何羽飞等: "基于测井多参数的特低渗储层产能快速预测", 《复杂油气藏》 *
刘晓红等: "低渗透储层产能预测的测井优化建模", 《西南石油大学学报(自然科学版)》 *
时卓等: "低渗透致密砂岩储层测井产能预测方法", 《测井技术》 *
郜燕群: "基于灰色综合关联度的科技金融有序性评价", 《喀什大学学报》 *
陈强等: "测井曲线灰关联度计算", 《测井技术》 *
高爱利: "南八区特高含水期压裂效果预测模型研究", 《中国优秀硕士学位论文全文数据库 工程科技I辑》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112949053A (en) * 2021-02-20 2021-06-11 东北石油大学 Dynamic prediction method for water injection rate of water injection oil well
CN114021804A (en) * 2021-11-02 2022-02-08 东北石油大学 Construction method of fault-lithology oil and gas reservoir oil and gas reserve prediction model
CN114021804B (en) * 2021-11-02 2022-07-19 东北石油大学 Construction method of fault-lithology oil and gas reservoir oil and gas reserve prediction model
CN115576028A (en) * 2022-12-01 2023-01-06 武汉盛华伟业科技股份有限公司 Geological feature layer prediction method and system based on support vector machine
CN115576028B (en) * 2022-12-01 2023-03-14 武汉盛华伟业科技股份有限公司 Geological feature layer prediction method and system based on support vector machine

Also Published As

Publication number Publication date
CN111784065B (en) 2021-04-16

Similar Documents

Publication Publication Date Title
CN111784065B (en) Oil well productivity intelligent prediction method based on grey correlation
CN108446797B (en) Method for predicting capacity of tight oil reservoir horizontal well at initial volume fracturing stage
CN110569566B (en) Method for predicting mechanical property of plate strip
CN114168906B (en) Mapping geographic information data acquisition system based on cloud computing
CN109214026A (en) Shale gas horizontal well initial-stage productivity prediction method
CN112901137A (en) Deep well drilling mechanical drilling speed prediction method based on deep neural network Sequential model
CN102799772A (en) Air quality forecast oriented sample optimization method
CN113865868B (en) Rolling bearing fault diagnosis method based on time-frequency domain expression
CN108897975A (en) Coalbed gas logging air content prediction technique based on deepness belief network
CN112016175A (en) Water supply pipe network pressure measuring point optimal arrangement method based on tree hierarchical clustering
CN114358434A (en) Drilling machine drilling speed prediction method based on LSTM recurrent neural network model
CN112861436A (en) Real-time prediction method for engine emission
CN113420795A (en) Mineral spectrum classification method based on void convolutional neural network
CN116703644A (en) Attention-RNN-based short-term power load prediction method
CN112836393B (en) Method for analyzing reservoir heterogeneity based on multi-scale entropy
CN114239397A (en) Soft measurement modeling method based on dynamic feature extraction and local weighted deep learning
CN116663126A (en) Bridge temperature effect prediction method based on channel attention BiLSTM model
CN112446168A (en) Effluent BOD concentration soft measurement method based on MIC and RBFNN
CN116842358A (en) Soft measurement modeling method based on multi-scale convolution and self-adaptive feature fusion
CN113973403B (en) Temperature-induced strain field redistribution intelligent sensing method based on structure discrete measurement point topology
CN112836274B (en) Data fusion method for tracing and auditing hidden engineering
CN110619167B (en) Oil-gas-water three-phase flow phase state critical point identification method based on cluster analysis
CN110189797B (en) Sequence error number prediction method based on DBN
Qian et al. Equity Evaluation of Global Countries based on the Expanded Theil index and Factor Analysis
CN111291490B (en) Nonlinear mapping intelligent modeling method for structure multi-scale heterogeneous response

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20221103

Address after: No. 19, Gangcheng Road, Dongying Port Economic Development Zone, Dongying City, Shandong Province 257237

Patentee after: Dongying Ruigang Pipeline Engineering Co.,Ltd.

Address before: 163000 99 Xuefu street, hi tech Industrial Development Zone, Daqing, Heilongjiang

Patentee before: NORTHEAST PETROLEUM University